6--数据增强（图片）

        图像增广在对训练图像进行一系列的随机变化之后，生成相似但不同的训练样本，从而扩大了训练集的规模。并且随机改变训练样本可以减少模型对某些属性的依赖，从而提高模型的泛化能力。

6.1 常用的图像增广方法

        左右翻转：torchvision.transforms.RandomHorizontalFlip(p=0.5)#默认以p=0.5的概率进行左右翻转

        上下翻转：transforms.RandomVerticalFlip(p=0.5)

        随机剪裁：transforms.RandomResizedCrop(size, scale=(0.1, 1), ratio=(0.5, 2))#

        改变颜色：torchvision.transforms.ColorJitter( brightness, contrast, saturation, hue)

        如需使用多种增广方式：

        torchvision.transforms.Compose([ aug1, aug2, aug3])

6.2 使用图像增广进行训练

        这里使用CIFAR-10数据集，是只包含imge-net中10个种类的子数据集。该数据集中对象的颜色和大小差异比fashion_mnist更明显。

!pip install git+https://github.com/d2l-ai/d2l-zh@release  # installing d2l
!pip install matplotlib_inline
!pip install matplotlib==3.0.0

%matplotlib inline
import torch
import torchvision
from torch import nn
from d2l import torch as d2l

all_images = torchvision.datasets.CIFAR10(train=True,root="data",download=True)
d2l.show_images([all_images[i][0] for i in range(12)],3,4,scale=1.5)

color_aug = torchvision.transforms.ColorJitter(brightness=0.5,contrast=0.5,saturation=0.5,hue=0.5)
#这里我使用了随机左右翻转和改变颜色
train_augs = torchvision.transforms.Compose(
    [torchvision.transforms.RandomHorizontalFlip(),
     color_aug,torchvision.transforms.ToTensor()])
test_augs = torchvision.transforms.Compose(
    [torchvision.transforms.ToTensor()])

def load_cifar10(is_train,augs,batch_size):
  dataset = torchvision.datasets.CIFAR10(
      root='data',train=is_train,transform=augs,download=True)#transform即用什么方法来进行图形增广
  dataloader = torch.utils.data.DataLoader(dataset,batch_size=batch_size)
  return dataloader

def train_batch_ch13(net,X,y,loss,trainer,devices):
  if isinstance(X,list):
    X = [x.to(devices[0]) for x in X]
  else:
    X = X.to(devices[0])
  y = y.to(devices[0])
  net.train()
  trainer.zero_grad()
  pred = net(X)
  l = loss(pred,y)
  l.sum().backward()
  trainer.step()
  train_loss_sum = l.sum()
  train_acc_sum = d2l.accuracy(pred,y)
  return train_loss_sum, train_acc_sum

def train_ch13(net, train_iter, test_iter, loss, trainer, num_epochs,
               devices=d2l.try_all_gpus()):
    """用多GPU进行模型训练"""
    timer, num_batches = d2l.Timer(), len(train_iter)
    animator = d2l.Animator(xlabel='epoch', xlim=[1, num_epochs], ylim=[0, 1],
                            legend=['train loss', 'train acc', 'test acc'])
    net = nn.DataParallel(net, device_ids=devices).to(devices[0])#并行
    for epoch in range(num_epochs):
        # 4个维度：储存训练损失，训练准确度，实例数，特点数
        metric = d2l.Accumulator(4)
        for i, (features, labels) in enumerate(train_iter):
            timer.start()
            l, acc = train_batch_ch13(
                net, features, labels, loss, trainer, devices)
            metric.add(l, acc, labels.shape[0], labels.numel())
            timer.stop()
            if (i + 1) % (num_batches // 5) == 0 or i == num_batches - 1:
                animator.add(epoch + (i + 1) / num_batches,
                             (metric[0] / metric[2], metric[1] / metric[3],
                              None))
        test_acc = d2l.evaluate_accuracy_gpu(net, test_iter)
        animator.add(epoch + 1, (None, None, test_acc))
    print(f'loss {metric[0] / metric[2]:.3f}, train acc '
          f'{metric[1] / metric[3]:.3f}, test acc {test_acc:.3f}')
    print(f'{metric[2] * num_epochs / timer.sum():.1f} examples/sec on '
          f'{str(devices)}')
    print('time:',timer.sum())

from torch.nn.init import xavier_uniform_
batch_size=256
devices = d2l.try_all_gpus()
net = d2l.resnet18(10,3)

def init_weights(m):
  if type(m) in [nn.Linear, nn.Conv2d]:
    nn.init,xavier_uniform_(m.weight)
net.apply(init_weights)

def train_with_data_aug(train_augs,test_augs,net,lr=0.001):
  train_iter = load_cifar10(True,train_augs,batch_size)
  test_iter = load_cifar10(False,test_augs,batch_size)
  loss = nn.CrossEntropyLoss(reduction='none')
  trainer = torch.optim.Adam(net.parameters(),lr=lr)
  train_ch13(net,train_iter,test_iter,loss,trainer,10,devices)

train_with_data_aug(train_augs,test_augs,net)

运行结果，我这里用了两种图像增广 咋测试精度还没有老师只用了一种的高呢？